Reputation: 31985
How can I install pandas dev version in my Python 2.7 (and OS X 10.9 and pip 1.4.1 for your information).
When I installed pandas using pip, the installed version is v.0.12 (latest stable version) for sure. However, when I cloned the repository from Github and installed it using setup.py
, the installed version is also v.0.12. So is it feasible to install v.0.13 under my system, and if it is, how can I do?
I don't want to use Scipy Superpack, which would install the dev version, since I like to install the dev version only on pandas, not on numpy, scipy, etc... as well.
I also want to install it under my Python 3.3 if it's feasible, but would be glad to accept it only on Python 2.7.
Thanks.
Upvotes: 2
Views: 3785
Reputation: 14144
Pandas v0.13.0 isn't out yet, and the GitHub repo still shows it as 0.12.0 within setup.py at the time of this writing:
MAJOR = 0
MINOR = 12
MICRO = 0
ISRELEASED = False
VERSION = '%d.%d.%d' % (MAJOR, MINOR, MICRO)
QUALIFIER = ''
FULLVERSION = VERSION
However, if you want to install a specific version of pandas (or any package), you can typically specify it:
$ pip install pandas==0.12.0
If you want to install straight from master on GitHub (which can be a scary beast), you can also do this directly:
$ pip install git+https://github.com/pydata/pandas.git
Hopefully you've installed setuptools and pip appropriately.
For python3, switch to pip3 (or pip-3.2, etc.). If you set python3 as the default within a virtualenv, pip is for python3:
$ mkvirtualenv --python=python3.2 tardis
...
(tardis) $ pip install pandas
Upvotes: 3